Re: Questions related to voice/sound recording in Debian.
There should be a number of solutions. If you build a
kernel and add the ALSA (Advanced Linux Sound Architecture)
project modules, the world of sound opens up very nicely. You
can adjust the soft controls on your sound card and such programs
as arecord and aplay work as advertised. The key is to know what
your sound card is, for sure, and build the right modules. When
you have done that, watch your system boot or use dmesg to make
sure that the sound system is happy in that it finds the sound
card and isn't complaining about things not being there, etc.
You can then youse amixer to learn what controls you have
and somewhat, what their capabilities are. They will vary from
one sound card to the other, but the important ones should be
there such as the microphone volume and high-level line inputs.
The sox application is a good program to have on your
system since it lets you do a lot of things to a sound file.
If you want to play with the most basic functions, you
can just use the raw 8-bit /dev/dsp directed to a file to get a
voice-grade audio stream. To play it back, do something like
cat filename >/dev/dsp
and you will get a roughly AM radio-quality sound as far as tone
quality.
You certainly need to get ALSA working, however, or your
sound efforts will be much more frustrating. You might get some
things to barely work, but not others.
I speak from bitter experience. I first started using
Linux in 2001 and had all kinds of trouble getting sound to work
predictably. I didn't know I needed a mixer and when I installed
one, it didn't always adjust what you'd expect it to. It's been
a while, but I remember that some settings also effected or
failed to effect the parameters they were supposed to be
changing. You might discover that you are unable to adjust the
output volume to anything but its power-up default setting, etc.
The ALSA drivers appear to accurately map the control registers
of the couple of different sound cards I use and the mixer
controls began to work just fine after ALSA.
One other thought. The slower your machine is, the
harder it will be to handle certain tasks such as playing video
or making 44.1 K samples per second audio recordings which are
the CD audio sampling rate. You may need to not run X or to not
perform certain I/O-intensive operations while recording or you
will find gaps or dropouts or speed variations in your recording.
It sometimes sounds like a tape recorder with a bad pinch roller
or other mechanical gremlin at work.
I have a 600-MHZ Pentium that generally does quite well
with sound, but starts croaking a bit or making bad recordings at
44100 samples per second if one does a big file transfer or the
find command which occupies the bus and prevents the audio data
from being moved fast enough so that some gets dropped. The more
RAM you have, the better and the faster your system is, the
better.
Martin McCormick WB5AGZ Stillwater, OK
Systems Engineer
OSU Information Technology Department Network Operations Group
Reply to: